{"id":"https://openalex.org/W3163560333","doi":"https://doi.org/10.1109/icassp39728.2021.9415039","title":"Internal Language Model Training for Domain-Adaptive End-To-End Speech Recognition","display_name":"Internal Language Model Training for Domain-Adaptive End-To-End Speech Recognition","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3163560333","doi":"https://doi.org/10.1109/icassp39728.2021.9415039","mag":"3163560333"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9415039","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9415039","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101749753","display_name":"Zhong Meng","orcid":"https://orcid.org/0000-0001-7814-5929"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Zhong Meng","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016279564","display_name":"Naoyuki Kanda","orcid":"https://orcid.org/0000-0002-8628-3288"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Naoyuki Kanda","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034136587","display_name":"Yashesh Gaur","orcid":null},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yashesh Gaur","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102376243","display_name":"Sarangarajan Parthasarathy","orcid":null},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sarangarajan Parthasarathy","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090360853","display_name":"Eric Sun","orcid":"https://orcid.org/0000-0001-8805-9864"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Eric Sun","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101607148","display_name":"Liang Lu","orcid":"https://orcid.org/0000-0003-4005-679X"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Liang Lu","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100329117","display_name":"Xie Chen","orcid":"https://orcid.org/0000-0001-5801-2571"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xie Chen","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100365053","display_name":"Jinyu Li","orcid":"https://orcid.org/0000-0002-1089-9748"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jinyu Li","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101928537","display_name":"Yifan Gong","orcid":"https://orcid.org/0000-0002-3912-097X"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yifan Gong","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":9,"corresponding_author_ids":["https://openalex.org/A5101749753"],"corresponding_institution_ids":["https://openalex.org/I1290206253"],"apc_list":null,"apc_paid":null,"fwci":5.3508,"has_fulltext":false,"cited_by_count":43,"citation_normalized_percentile":{"value":0.96292823,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"7338","last_page":"7342"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9968000054359436,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.993399977684021,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7954342365264893},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6893784999847412},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5623875260353088},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.560845136642456},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5328776240348816},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.478958398103714},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4553471505641937},{"id":"https://openalex.org/keywords/modular-design","display_name":"Modular design","score":0.4517022669315338},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.4360159635543823},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.4277839660644531},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.41801929473876953},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.08333170413970947}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7954342365264893},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6893784999847412},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5623875260353088},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.560845136642456},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5328776240348816},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.478958398103714},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4553471505641937},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.4517022669315338},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.4360159635543823},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.4277839660644531},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.41801929473876953},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.08333170413970947},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9415039","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9415039","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":65,"referenced_works":["https://openalex.org/W854541894","https://openalex.org/W1494198834","https://openalex.org/W1513820424","https://openalex.org/W1828163288","https://openalex.org/W1915251500","https://openalex.org/W1922655562","https://openalex.org/W1989549063","https://openalex.org/W1993409002","https://openalex.org/W2010362084","https://openalex.org/W2080005694","https://openalex.org/W2094147890","https://openalex.org/W2127141656","https://openalex.org/W2160815625","https://openalex.org/W2293634267","https://openalex.org/W2327501763","https://openalex.org/W2402040300","https://openalex.org/W2510867321","https://openalex.org/W2515801922","https://openalex.org/W2545177271","https://openalex.org/W2577366047","https://openalex.org/W2584667682","https://openalex.org/W2636483419","https://openalex.org/W2769025471","https://openalex.org/W2795867901","https://openalex.org/W2796339975","https://openalex.org/W2891980359","https://openalex.org/W2911629330","https://openalex.org/W2936252403","https://openalex.org/W2939164678","https://openalex.org/W2962784628","https://openalex.org/W2962824709","https://openalex.org/W2963970535","https://openalex.org/W2972621414","https://openalex.org/W2972818416","https://openalex.org/W2973138437","https://openalex.org/W2987019345","https://openalex.org/W3006806348","https://openalex.org/W3007328579","https://openalex.org/W3008008574","https://openalex.org/W3008037978","https://openalex.org/W3011956499","https://openalex.org/W3015194534","https://openalex.org/W3015915933","https://openalex.org/W3016234571","https://openalex.org/W3094667432","https://openalex.org/W3095311338","https://openalex.org/W3106526813","https://openalex.org/W3112742522","https://openalex.org/W3152221657","https://openalex.org/W3162639446","https://openalex.org/W6623517193","https://openalex.org/W6638575559","https://openalex.org/W6638749077","https://openalex.org/W6640059789","https://openalex.org/W6640090968","https://openalex.org/W6712847557","https://openalex.org/W6728910023","https://openalex.org/W6732447497","https://openalex.org/W6732862412","https://openalex.org/W6740080312","https://openalex.org/W6770245836","https://openalex.org/W6779264309","https://openalex.org/W6785025786","https://openalex.org/W6787533431","https://openalex.org/W6793527289"],"related_works":["https://openalex.org/W2594897229","https://openalex.org/W2151348424","https://openalex.org/W4221142855","https://openalex.org/W2050138804","https://openalex.org/W2944691285","https://openalex.org/W4398173524","https://openalex.org/W4290708361","https://openalex.org/W2129812225","https://openalex.org/W26527944","https://openalex.org/W2523799048"],"abstract_inverted_index":{"The":[0],"efficacy":[1],"of":[2,124,137],"external":[3,50,140],"language":[4,24],"model":[5,25,80,92,112],"(LM)":[6],"integration":[7,136],"with":[8,48,113,143,156,174],"existing":[9,100],"end-to-end":[10],"(E2E)":[11],"automatic":[12],"speech":[13],"recognition":[14],"(ASR)":[15],"systems":[16],"can":[17],"be":[18],"improved":[19],"significantly":[20],"using":[21],"the":[22,33,40,45,49,57,78,84,90,108,125,138],"internal":[23,34,63,72,85,127],"estimation":[26],"(ILME)":[27],"method":[28,67],"[1].":[29],"In":[30],"this":[31],"method,":[32],"LM":[35,51,64,73,86,97],"score":[36,41,47],"is":[37],"subtracted":[38],"from":[39,170],"obtained":[42],"by":[43,75],"interpolating":[44],"E2E":[46,79,91,111,172],"score,":[52],"during":[53],"inference.":[54],"To":[55],"improve":[56],"ILME-based":[58,157],"inference,":[59],"we":[60],"propose":[61],"an":[62,70],"training":[65,115,173],"(ILMT)":[66],"to":[68,93,132,161],"minimize":[69],"additional":[71],"loss":[74],"updating":[76],"only":[77],"components":[81],"that":[82],"affect":[83],"estimation.":[87],"ILMT":[88,155],"encourages":[89],"form":[94],"a":[95,120,133],"standalone":[96],"inside":[98],"its":[99],"components,":[101],"without":[102],"sacrificing":[103],"ASR":[104],"accuracy.":[105],"After":[106],"ILMT,":[107],"more":[109,121,134],"modular":[110],"matched":[114],"and":[116,129,150,163,180],"inference":[117,158],"criteria":[118],"enables":[119],"thorough":[122],"elimination":[123],"source-domain":[126],"LM,":[128],"therefore":[130],"leads":[131],"effective":[135],"target-domain":[139],"LM.":[141],"Experimented":[142],"30K-hour":[144],"trained":[145],"recurrent":[146],"neural":[147],"network":[148],"transducer":[149],"attention-based":[151],"encoder-":[152],"decoder":[153],"models,":[154],"achieves":[159],"up":[160],"31.5%":[162],"11.4%":[164],"relative":[165],"word":[166],"error":[167],"rate":[168],"reductions":[169],"standard":[171],"Shallow":[175],"Fusion":[176],"on":[177],"out-of-domain":[178],"LibriSpeech":[179],"in-domain":[181],"Microsoft":[182],"production":[183],"test":[184],"sets,":[185],"respectively.":[186]},"counts_by_year":[{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":16},{"year":2022,"cited_by_count":13},{"year":2021,"cited_by_count":4}],"updated_date":"2026-04-03T22:45:19.894376","created_date":"2025-10-10T00:00:00"}